M. de Beanmout on Ihe Olacitr Theory. Ill 



this liquefaction, about G millimetres of water, we should not 

 be very far from the truth for any given point. 



This influx of heat proceeding from the interior of the 

 earth, arrives at the bed of glaciers as at the bed of the sea 

 and of lakes, and, in general, at all points of the rocky crust 

 of the earth. Having reached the bed of a glacier, it conducts 

 itself differently, according to circumstances, as I have already 

 remarked in a note read to the Philomathie Society on the 6th 

 June 1836. (See VInstUnt, vol. iv. p. 192, No. 162, June 

 15th, 1836.) The flow of heat may traverse the entire gla- 

 cier, and then become dissipated at its surface; or it may 

 stop at the bed of the glacier, and be there entirely employed 

 in melting the ice; or, more generally, it may become divided 

 into two portions, of which the one is employed in melting the 

 ice, and the other traverses the ice, and is dissipated at the 

 sui'face by radiation, by contact with the air, &;c. 



Hence it results that the )/ia.nr?inm quantity of water which 

 can result from the action of central heat on the ice and 

 snow distributed over the surface of the earth, is represented 

 by a sheet of water six millimetres in thickness, having the 

 same extent as that ice and that snow, and that the maximum 

 quantity which can be produced in a month, is represented by 

 a sheet of water half a millimitre in thickness, a quantity cor- 

 responding with that produced by a very small fall of rain. 



The quantity of water residting from the liquefaction 

 caused by the sun, and by atmospherical actions, is incompar- 

 ably greater. 



In the physical atlas of Berghaus, the quantity of water 

 which falls annually on the elevated portions of the Alps, in 

 the state of rain, hail, or snow, is estimated at thirty-five inches, 

 or 947 millimetres ; the snows and the glaciers of the Alps 

 having remained for many ages in a state almost stationary, 

 but more retrograde than progressive, it must necessarily be 

 the case that the quantity of water which flows from them 

 annually (apart from the evaporation) must be equivalent to 

 that which falls in all forms ; this quantity ought even to 

 exceed, relatively to the surface really covered by permanent 

 snow or ice, the proportion stated above, because, the slopes 

 which are too rapid for the adherence of snow, throw off all 



